Chapter 1
Consuming Research
Consuming research occurs in your everyday life
When you read a book, there is substantial research that goes into the development of the book.
D-Day, June 6th, 1944 by Stephen Ambrose (non-fiction)
The Da Vinci Code by Dan Brown (fiction)
Parenting, “how-to,” and other forms of books are based on research!
Favorite Netflix documentary? Based on research
Deciding what brand of shampoo to buy? You do your research (probably)
Active ingredients, what your barber/salon uses, asking friends or relatives
Why Learn About Research Methods
As future family practitioners (ex., therapists, teachers), you will use research to inform what interventions you use.
You make better and more informed decisions (increase positive outcomes and decrease negative outcomes)
Research informs public policy
Policy is crafted in large part based on known researchers
Doctors can receive kickbacks (cash) from pharmaceutical companies for prescribing the company’s medication a certain number of times
The more you prescribe the medication, the more money the doctor makes
It will directly affect you
Medications created by Pfizer, Eli Lily and other companies are tested, but would you trust a company that made a medication AND they were the only ones who tested its efficacy?
Companies invest billions into the development of medication, so aren’t they investing in getting it to market and it working? Are they unbiased?
Seeks to help us answer the following questions
“How do you know?”
“What do you mean by that?”
“How sure are you that it is not ______?”
“Does this knowledge apply to _________?”
Vioxx
Made by Merck, was an anti-pain medication used for arthritis and made about 2.5 billion per year (20 million people)
140,000 or more people developed heart disease after taking Vioxx
Vioxx and the FDA ignored harmful side effects for years
Reports on Vioxx were sponsored by Vioxx (spent 21 million dollars to publish the report)
The report said that Merck acted in good faith
Settled for 4.8 billion dollars
Fen-Phen
Weight loss drug made by Wyeth
6.5 million used Fen-Phen
Users developed severe lung issues (pulmonary hypertension)
Was recalled in 1997, but the company was undercounting the number of pulmonary hypertension cases, starting in 1994
Cost an estimated 21 billion in legal defense and damages
Ways of Knowing
Intuition
Anecdotal evidence where we can find explanations for why things happened (but after the fact)
Knowledge is based individial expereinces and generalized to the world around us
Knowledge about a very specific event or process in our own lives is generalized to our lives more generally
We study for a test, and we end up doing well → I’m good at studying
A fight with your partner over doing the dishes → My partner doesn’t care about the things that I value
When two things happen that are out of the norm, either simultaneously or in close succession to one another, we can believe there is a relationship between the two events when there isn’t one
Illusory correlation (Individual)
Spurious correlation (Statistics)
Examples:
Doing a specific therapeutic technique with one person and it being highly effective → believing it will be effective for all clients
We absolutely nailed the first exam in research methods with little studying, so we conclude that we don’t need to study much for all the exams.
Rationalizations: we develop justifications for why we behaved a certain way; most often, these justifications are inapplicable or not appropriate
Often, to escape feelings of discomfort, shame, and guilt
Examples
We don’t get into graduate school, so we say we didn’t really want to go to grad school.
We explain that our criticism of our partner is because they criticised us first, and if they hadn’t done it, then we wouldn’t have
Pros
Easily accessible
Informed by our own experiences, which may be more relevant than others’ experiences
Why others’ advice is actually very rarely helpful
Generally, intuition is easy to create and allows us to avoid discomfort
Cons
We all have biases (ex., confirmation bias)
Example: One partner in couples therapy asks the therapist, “Aren’t they being an *******!?”
Post-hoc: reasons are created after the fact
Not systematic - our conclusions can be confounded by other processes that we don’t think about
Example: Our partner comes home, and we greet them, but they ignore us. We could react based on that single piece of information (angry), or we could consider this information in the context of their behavior historically (unlike them → they had a bad day)
Authority
Those in authority often persuade humans
We adopt many of the beliefs, attitudes, and behaviors of our family members (especially parents)
We listen to celebrities when they endorse something → this is why celebrities do commercials.
We trust those with whom we have a close relationship, those who are well-known, those who are persuasive, trustworthy, or respectable.
You likely have friends or family members whom you would take advice from, and those who you would NEVER take advice from
When trusting an authority, you are placing your faith in that person’s knowledge/endorsement.
Reject the notion that we accept the positions of others as the TRUTH
Blind faith goes against the scientific process
The problem with authority is the blind acceptance, not with the learning/information itself.
Trust but verify
Pros
We don’t have to think! We absorb the information, and it becomes our own
We learn a TON of good information from authority figures
A lot of information that we learn from authorities is fundamental and not consequential in the grand scheme of things
Cons
We don’t have to think! We absorb the information, and it becomes our own
The information we learn may not be a true authority and could be their own intuition
Empiricism
Many Scientists (and some laypeople) recognize that their ideas may be wrong.
Anecdotal, intuition may not apply to everyone else
Avoid blind acceptance of authority
Knowledge comes from systematic observation
The scientific method provides a way to gain knowledge using a specific set of principles in gathering data.
Scientific Method
Engage in the scientific method to understand what precisely is happening, identifying the best way to gather data, and making appropriate and accurate conclusions.
Principles of the scientific method
Data plays a critical role
Scientists are not alone
Science is adversarial
Data
For scientists, knowledge is based on systematic (ex., intentional) observation of a phenomenon of interest and generating conclusions based on the data.
We develop theories based on the data (or use data to test theories) and refine our understanding based on that data.
Best practice is to repeatedly gather data
Use high-quality practices/methods to gather data
Science is a Team Sport
Knowledge is cumulative, where scientific ideas created by one person are subsequently examined by other scientists
Our ideas CAN be biased due to research questions based on our intuition or authority.
Require an independent test or evaluation.
Science also requires researchers to work together - the more heads that are together, the better ideas will lead to better ideas.
Science is Adversarial
Scientific debate is the bedrock of advancement.
Developing a logical and coherent rationale for why a phenomenon exists that is expected to be criticized by peers.
Rejection is a part of science
Rejection is also NOT a personal attack
Science is Peer Reviewed
Science is conducted by experts - but also evaluated (often anonymously) by other experts (ex., peer reviews)
Without the peer review, research is nothing more than authority-based knowledge
Skepticism
The more you learn about science, the more you become skeptical of research findings
How much should we believe that someone who says they are a scientist is an expert?
Not all science is science
Pseudoscience occurs when people use scientific language or demonstrations to substantiate claims of proof, but not based on scientific evidence
Pseudoscience
Claims that appear to be scientifically drawn, but are instead based on untestable hypotheses, are subject to confirmation bias, lack of peer-reviewed findings, and the data used are proprietary and not available for independent examination
Claims are not based on the scientific method
Reliance on anecdotes (ex., a selection effect)
Example: Astrology
Identifying Pseudoscience
Claims that are untestable and therefore cannot be proven correct or incorrect
Claims that are contradictory, confusing, vague, unnecessarily complex, or biased
Evidence based on anecdotes or testimonials
Notice how Chick-fil-A commercials say real customers paid for the testimonials
Use experts with not vague credentials to justify expertise
Claims that only have positive evidence, ignoring critical evidence
The science cannot be independently verified - “trust me, bro.”
Goals of Behavioral Science
Description
Aims to quantify the nature of a specific construct that is directly observed, including how common a phenomenon is, what percentage of people experience a phenomenon, similarity and differences among people
Examples
64% of adults in the United States experienced at least one adverse childhood experience
75% of black adults experience at least one ACE, and 57% of white adults experience at least one ACE
The average household income in the United States is $80,000, but only 22% of Americans make at least $80,000
Prediction
After we establish that our independent and dependent variables are correlated with each other (positively or negatively), we want to make a prediction
Correlated: tendency for two variables to co-occur
Prediction: one variable precedes / predicts the other; aspect of temporal precedence
Childhood abuse predicts an increase in the odds of having major depressive disorder
In the social sciences, prediction is probabilistic rather than causal
Many people who experience ACEs do not have major depressive disorder
Determining Causes
Scientists are interested in determining what causes a specific behavior
This is used in the prevention of effects
Example: Pre-marital therapy reduces the likelihood of divorce
In our example, we have been thinking about the consequences of child abuse (depression), but what causes child abuse?
Socioeconomic status, parental psychopathology
Behavior has numerous causes (making it both challenging and interesting to study)
Other fields (engineering) can make exceptionally strong causal claims because there are relatively fewer predictors.
In the social sciences, if we can explain 20% of the person-to-person differences in a behavior, we’re doing well, but in some engineering, if they explain 99% of thhe conditon to condition differences, they’re upset
When determining causes, we must establish temporal precedence
The cause must precede the effect
Why?
The co-occurrence (covariation) of two variables can be a coincidence!
The relationship can be confounded
An unmeasured variable is a common cause of both variables
There is an association between Ice Cream Consumption and Violence
It could also be that there is a covariation between two variables, but there are alternative explanations that take away the covariation between two variables
Example: Socioeconomic status may cause childhood maltreatment, but when accounting for parental trait anger, the association goes away